Overview

Smoking is a major public health concern and is associated with numerous health risks. Automatically identifying smoking behavior from images can support applications such as public health monitoring, workplace safety compliance, and content moderation systems. In this challenge, your task is to build a computer vision model that can determine whether an image contains evidence of smoking activity.

You will work with a labeled image dataset consisting of smoker and non-smoker examples, providing an opportunity to practice essential Computer Vision (CV) skills such as image preprocessing, feature extraction, and deep learning-based classification. This challenge is beginner-friendly and designed to help participants gain hands-on experience in building end-to-end image classification pipelines, from data exploration and model training to prediction and evaluation.

Practice Skills

In this challenge, you will gain hands-on experience with:

• Python
• Image Preprocessing
• Feature Extraction
• Deep Learning
• Image Classification
• Model Evaluation

Evaluation

Goal

Train a classification model that predicts whether each image in the test set contains smoking activity (Smoker) or does not contain smoking activity (Non-Smoker).

Metric

Submissions are evaluated on Accuracy:

$\text{Accuracy}=\frac{\text{Correct Predictions}}{\text{Total Predictions}}$

Submission

Format

A CSV with two columns:

• id: The image filename.
• label: The predicted label for the corresponding image (0 = non-smoker, 1 = smoker).

id	label
6138699878db49328e6f59e6337cd7ee.jpg	0
7d16690af8d0420f9d6b9647354f0300.jpg	1

See the submission guide for upload instructions.

Tracks

The challenge accepts two submission tracks:

• Public: predict on the public test set and upload the resulting CSV.
• Private: upload your model. It is scored on a held-out private test set. Your code must recursively scan the entire data directory.